Yes / Individual radial frequency (RF) patterns are generated by modulating a circle's radius as a sinusoidal function of polar angle and have been shown to tap into global shape processing mechanisms. Composite RF patterns can reproduce the complex outlines of natural shapes and examining these stimuli may allow us to interrogate global shape mechanisms that are recruited in biologically relevant tasks. We present evidence for a global shape aftereffect in a composite RF pattern stimulus comprising two RF components. Manipulations of the shape, location, size and spatial frequency of the stimuli revealed that this aftereffect could only be explained by the attenuation of intermediate-level global shape mechanisms. The tuning of the aftereffect to test stimulus size also revealed two mechanisms underlying the aftereffect; one that was tuned to size and one that was invariant. Finally, we show that these shape mechanisms may encode some RF information. However, the RF encoding we found was not capable of explaining the full extent of the aftereffect, indicating that encoding of other shape features such as curvature are also important in global shape processing. / This research was supported by a Biotechnology and Biological Sciences Research Council (BBSRC) grant #BB/L007770/1.

Non-rigid point set registration of significantly uneven deformations is a challenging problem for many applications such as pose estimation, three-dimensional object reconstruction, human movement tracking. In this dissertation, we present a novel probabilistic non-rigid registration method to align point sets with significantly uneven deformations by enforcing constraints from corresponding key points and preserving local neighborhood structures. The registration method is treated as a density estimation problem. Incorporating correspondence among key points regulates the optimization process for large, uneven deformations. In addition, by leveraging neighborhood embedding using Stochastic Neighbor Embedding (SNE) as well as an alternative means based on Locally Linear Embedding (LLE), our method penalizes the incoherent transformation and hence preserves the local structure of point sets. Also, our method detects key points in the point sets based on geodesic distance. Correspondences are established using a new cluster-based, region-aware feature descriptor. This feature descriptor encodes the association of a cluster to the left-right (symmetry) or upper-lower regions of the point sets. We conducted comparison studies using public point sets and our Human point sets. Our experimental results demonstrate that our proposed method successfully reduced the registration error by at least 42.2% in contrast to the state-of-the-art method. Especially, our method demonstrated much superior performance in the case of a large degree of deformations. Experimental results show more than 30% improvements when key point correspondence is used. Our study shows that the influence of the global constraint (key point correspondences) is greater than that of the local constraint. Our analysis of using incorrect key point correspondence reveals the sensitivity of the proposed method. Given erroneous but symmetric correspondence, our method was able to produce fairly good results. In addition, our study on time reports a competitive computational efficiency of the proposed method.

<p>This thesis presents an approach to free-form surface manipulations, which conceptually improves an existing CAD system that constructs surfaces by smoothly interpolating a network of intersecting curves. There are no regularity requirements on the network, which already yields superior modelling capabilities compared to systems that are based on industry-standard NURBS surfaces.</p><p>Originally, the shape of such a surface can be modified only locally by manipulating a curve in the network. In this process there is an inherent danger that the curve is being pulled away from intersections that it has with other curves. When this happens, the network is invalidated as a surface representation, and many curves may have to be adjusted to restore network consistency and surface quality. This thesis contributes a method that solves these problems by propagating changes that are made in one curve to curves in its vicinity. How and to what extent curves react to changes is controlled by two parameters that can be varied along the curve that is being manipulated. Any curve may be constrained in one or more degrees of freedom. The integrity of the curve network is implicitly conserved, as well as the geometric continuity of the surface.</p><p>The result is a tool for the modification of curve-interpolating surfaces, which can easily be applied to large areas on models with any level of detail. This allows designers to concentrate on the creative process, rather than on planning chains of actions. They can explore different design variations, optimise shapes further, and generally be more productive.</p> / <p>Dette doktorgradsarbeidet presenterer en fremgangsmåte for formgivning og modifisering av datamaskinbaserte, skulpturerte flater. Metoden forbedrer et eksisterende system for data-assistert konstruksjon (DAK) som bygger dobbeltkrummede flater ved å interpolere et nettverk av skjærende kurver. Nettverket trenger ikke være regelmessig, noe som allerede gir bedre modelleringsmuligheter sammenliknet med systemer som er basert på standard NURBS flater.</p><p>En slik flate kan opprinnelig bare endres lokalt ved å dra i en kurve. I denne prosessen er det fare for at kurven blir dratt fra skjæringspunkter den har med andre kurver. Hvis dette skjer, representerer ikke nettverket en flate lenger, og mange kurver må justeres for å få tilbake integriteten i nettverket og kvaliteten i formen. Denne avhandlingen bidrar med en metode som løser disse problemene ved å spre endringer som blir gjort i en kurve til andre kurver i nærheten. Hvordan og i hvilken utstrekning kurvene reagerer på endringen styres av to parametre som kan varieres langs kurven som blir endret. Enhver kurve kan låses i en eller flere frihetsgrader. Integriteten til nettverket samt glattheten i formen blir bevart automatisk.</p><p>Resultatet er et redskap for modifikasjon av kurve-interpolerende flater som med letthet kan brukes på større områder av modeller med hvilken som helst grad av detalj. Dette gir designere muligheten til å konsentrere seg om det kreative, istedenfor å planlegge handlingsrekker. De kan utforske forskjellige designvariasjoner, optimalisere former ytterligere, og i det hele tatt være mer produktive.</p>

This thesis presents an approach to free-form surface manipulations, which conceptually improves an existing CAD system that constructs surfaces by smoothly interpolating a network of intersecting curves. There are no regularity requirements on the network, which already yields superior modelling capabilities compared to systems that are based on industry-standard NURBS surfaces. Originally, the shape of such a surface can be modified only locally by manipulating a curve in the network. In this process there is an inherent danger that the curve is being pulled away from intersections that it has with other curves. When this happens, the network is invalidated as a surface representation, and many curves may have to be adjusted to restore network consistency and surface quality. This thesis contributes a method that solves these problems by propagating changes that are made in one curve to curves in its vicinity. How and to what extent curves react to changes is controlled by two parameters that can be varied along the curve that is being manipulated. Any curve may be constrained in one or more degrees of freedom. The integrity of the curve network is implicitly conserved, as well as the geometric continuity of the surface. The result is a tool for the modification of curve-interpolating surfaces, which can easily be applied to large areas on models with any level of detail. This allows designers to concentrate on the creative process, rather than on planning chains of actions. They can explore different design variations, optimise shapes further, and generally be more productive. / Dette doktorgradsarbeidet presenterer en fremgangsmåte for formgivning og modifisering av datamaskinbaserte, skulpturerte flater. Metoden forbedrer et eksisterende system for data-assistert konstruksjon (DAK) som bygger dobbeltkrummede flater ved å interpolere et nettverk av skjærende kurver. Nettverket trenger ikke være regelmessig, noe som allerede gir bedre modelleringsmuligheter sammenliknet med systemer som er basert på standard NURBS flater. En slik flate kan opprinnelig bare endres lokalt ved å dra i en kurve. I denne prosessen er det fare for at kurven blir dratt fra skjæringspunkter den har med andre kurver. Hvis dette skjer, representerer ikke nettverket en flate lenger, og mange kurver må justeres for å få tilbake integriteten i nettverket og kvaliteten i formen. Denne avhandlingen bidrar med en metode som løser disse problemene ved å spre endringer som blir gjort i en kurve til andre kurver i nærheten. Hvordan og i hvilken utstrekning kurvene reagerer på endringen styres av to parametre som kan varieres langs kurven som blir endret. Enhver kurve kan låses i en eller flere frihetsgrader. Integriteten til nettverket samt glattheten i formen blir bevart automatisk. Resultatet er et redskap for modifikasjon av kurve-interpolerende flater som med letthet kan brukes på større områder av modeller med hvilken som helst grad av detalj. Dette gir designere muligheten til å konsentrere seg om det kreative, istedenfor å planlegge handlingsrekker. De kan utforske forskjellige designvariasjoner, optimalisere former ytterligere, og i det hele tatt være mer produktive.